Research On Low-cost Indoor Positioning Method Based On Ranging In Wireless Sensor Networks

Wireless sensor networks(WSN)is a good solution for indoor localization,because of its low cost,flexible deployment,robust performance and so on.RSSI-based localization and CSI-based localization are WSN-based localizations,they use the RSSI and CSI from the anchors respectively,which can achieve higher location accuracy with lower hardware cost.So RSSI-based localization and CSI-based localization have a Good market application prospect.This thesis mainly studies these two kinds localization with theoretical analysis and experimental analysis.The main work is as follows:(1)RSSI values can be easily influenced by multipath fading,noise,temperature and other environmental parameters.This decreases the accuracy and stability of estimated distance.To improve localization accuracy,this thesis proposes a multiplicative distance-correction factor to counteract the inaccuracy of estimated distance.In the same indoor environment,the product of this correction factor and estimated distance is regarded as a good approximation of real distance between unknown node and an anchor node.Based on multiplicative distance-correction factor,two location estimation algorithms,MDCF-Grid and MDCF-PSO are proposed in this thesis.According to the experimental results,MDCF-Grid and MDCF-PSO have higher location accuracy and stability than the traditional location estimation algorithm.(2)This thesis analyses RSSI-based and CSI-based localization in theory.According to the subcarrier's impulse response,different ranging models are built to convert CSI to estimated distance based on the difference in requirements and specific scenario.Besides,low bound of estimated location is computed based on different ranging models respectively.(3)A lot of experiments have been done to achieve enough RSSI values.According to the analysis of RSSI values in different scenario,different kinds of RSSI processing are proposed in this thesis based on the problems of these values.A kind of predictive RSSI processing method is proposed especially.This method can counteract the error caused by time delay,which is related with unknown node's moving.